Call for Testing

November 4, 2009 at 10:39 am 32 comments

bmesh subsurf ngon

BMesh is now in the alpha testing stage.  It’s reached the point where it needs user testing to further progress and identify bugs.

You can find builds on graphicall.org, and report bugs on redmine.  Please use the redmine bug tracker!  Feel free to leave comments in the blog, of course.

Build Links

Please use a build no older then November 4th.

Linux SuSe: http://www.graphicall.org/builds/builds/showbuild.php?action=show&id=1164

Win32: http://graphicall.org/builds/builds/showbuild.php?action=show&id=1165

QuickStart

Dissolve is xkey->dissolve, and vertex connect is ykey.  Those two are used a lot, so it’s a good idea to familiarize yourself with them.  Subsurf now works with ngons, which allows for all sorts of cool modeling tricks (and to get rid of the ngons you simply apply the first level of subsurf).

I’ve also made a start of feature notes on blender.org.

Crash Recovering

Blender will now save recovery files on crashing.  So if blender crashes, look for a [filename].crash.blend file.   Note that the recovery file will not include changes made in editmode, so I’d suggest cycling through editmode/objectmode every once in a while.

Known Issues

  • GLSL draw mode doesn’t work.
  • Edge Slide doesn’t work.
  • Particles are buggy.
  • MultiRes hasn’t been brought back yet.
  • Sculpting may not work.
  • The vkey rip tool isn’t quite finished yet.

Notes

Not everything works yet.  Some parts need attention by their maintainers, e.g. particles and MultiRes, and are things I shouldn’t really be dealing with (meddling in 😉 ) myself.  And I’m sure there’s a great many bugs I’ve not uncovered yet myself.

On the other hand, things are fairly functional.  Most tools have been ported over.  Only a few are missing.

Donate!

Please remember to donate, even small contributions help.  Donations are what allow me to spend as much time on bmesh as I do.

Advertisements

Entry filed under: Uncategorized.

Status Update Donations

32 Comments Add your own

  • […] created a win32 build of joeedh’s bmesh branch. See his call for testers for more info (note the redmine link for bug […]

    Reply
  • 2. Paweł Łyczkowski  |  November 4, 2009 at 12:01 pm

    Yeeeeeees!!! Starting to test.

    Reply
  • 4. ijstaart  |  November 4, 2009 at 12:38 pm

    I haven’t tried the build yet (but I will), but will there also be an edge connect and face connect? (face connect being more like a bridge tool).

    It works that way in Silo and it’s really convenient to have all that functionality in one function / under one key 🙂

    Thank you for doing the great work on giving Blender an awesome mesh editing system!

    Reply
    • 5. joeedh  |  November 4, 2009 at 12:43 pm

      Edge connect is edge subdivide. A bridge tool would be nice, not sure when I’ll get to it though.

      Reply
  • 6. Justin  |  November 4, 2009 at 12:51 pm

    funny, I just posted on the previous post about making an update…lol…Good Stuff!

    Reply
  • 7. Paweł Łyczkowski  |  November 4, 2009 at 1:46 pm

    If vertex slide would be easy to do, I’d love to see this feature implemented.

    Reply
    • 8. Paweł Łyczkowski  |  November 4, 2009 at 2:29 pm

      Also, when using multiple loopcut (ctrl-r, mouse wheel up), a sliding to the edges and back would be great…

      Reply
  • 9. Kharon  |  November 4, 2009 at 5:06 pm

    I’d love to test if there was only a Mac build available — I mean, I will be testing over the weekend once I’m at work, but at home I have no Windows box, only a Mac Intel setup. If anybody knows of, or can produce, a current BMesh build for Mac Intel, I’ll do a lot more testing.

    Reply
  • 10. michal Bradwolff  |  November 4, 2009 at 5:43 pm

    Is there a Mac OS version of it?
    10.5 or better an 10.6 version.

    Reply
  • 11. Ed  |  November 4, 2009 at 6:39 pm

    Hi, I’m having a problem understanding delete. I was under the impression that we could now arbitrarily delete edges and vertices, much like maya/max/silo/xsi?

    Running delete on an edge for instance, will still remove the surrounding faces. And running dissolve on most edges doesn’t seem to do anything.

    Reply
  • 12. joeedh  |  November 4, 2009 at 6:49 pm

    @Ed: Could you describe what cases are failing for edge dissolve? Dissolve should work flawlessly in vertex and face selection mode, but edge selection mode is still a bit iffy (I need to refactor the code that vertex, edge and face dissolve all share so it works better).

    Dissolve is what’s supposed to nondestructively remove geometry; delete works as it did before (to avoid pissing people off).

    Reply
    • 13. Ed  |  November 5, 2009 at 12:52 am

      Sorry, I’ll have to play with it some more. I had fiddled around and made a mesh that wouldn’t let me dissolve anything… but now it’s working 🙂 I will try to recreate my issue

      Reply
  • 14. AlphaSite  |  November 4, 2009 at 10:43 pm

    Having issues compiling to mac, any cnahce of a quick fix up or tip? I tied deleting all of these before but came up with another unrelated error, i don’t have a log of that one but here is one of the standard error:-

    CMake Error: Error in cmake code at
    /users/user/bf-blender/blender_bmesh/blender/source/creator/CMakeLists.txt:110:
    Parse error. Expected a command name, got unquoted argument with text “<<<<<<<".

    Reply
    • 15. John Croisant  |  November 6, 2009 at 8:24 pm

      Oops, looks like some merge conflicts accidently got committed. They are breaking CMake compile for me too (on Gentoo Linux).

      I would suggest trying compiling with scons as a workaround until that gets fixed.

      Reply
  • 16. vykelt  |  November 5, 2009 at 1:20 am

    oh wow. This version works so much better then earlier bmesh builds. I haven’t been able to break it yet. I’ll keep trying, but great work. so far. still can’t wait to see the end result of all your hard work. thanks joe

    Reply
  • 17. Artintal  |  November 5, 2009 at 2:57 am

    I am not really into 3d modelling as much as I am into watching the development so i cant really test it but I gotta say “Thank you”. What you are doing, more or less by yourself, is amazing!

    Reply
  • 18. dsavi  |  November 6, 2009 at 6:09 am

    The Linux build you linked to does not work on Ubuntu. (Karmic Koala)

    “./linux2/blender: error while loading shared libraries: libpython3.1.so.1.0: cannot open shared object file: No such file or directory”

    I don’t know a lot about this, but if it’s trying to get Python 3.1 from outside of Blender, I know that it isn’t there. (Ubuntu doesn’t have Python >= 3.x in the repositories yet) If it’s looking for bundled Python, well, then I don’t have any idea. I would compile it myself (And redistribute it) if I didn’t want to mess up my installation (Requires installing some weird packages from Debian testing that break some Python programs)

    Reply
    • 19. dsavi  |  November 6, 2009 at 6:38 am

      In addition to my previous post, it seems that Ubuntu does have libpython-3.1, but that’s apparently not enough to run Blender with Bmesh. I get the following error (After installing libpython-3.1)
      ./linux2/blender: symbol lookup error: ./linux2/blender: undefined symbol: PyUnicodeUCS2_FromString

      I hardly know Python so that doesn’t mean a lot to me. 😛

      Reply
      • 20. joeedh  |  November 6, 2009 at 1:34 pm

        I have no idea, heh. Sounds like the ubuntu py libs weren’t built right, but I don’t know what to do about that. Any linuxers here who can help?

      • 21. mbue  |  November 6, 2009 at 8:54 pm

        I actually built the python 3.1 lib myself (and use it with blender2.5 successfully), but could not get the bmesh branch to compile with it 😦

        Defining the correct python path in user-config.py (scons) like i did for 2.5 trunk doesn’t change anything.

        Also: cmake&make spits out errors about not finding bmesh.h (and a lot of other stuff I didn’t bother to investigate further yet. e.g. some problem with AUD_C-API.h which is not related to bmesh)
        I’m stumped right now on how to solve this.

        Another thing: I actually had to edit/remove the file
        blender/source/creator/CMakeLists.txt
        because it had some DIFF syntax in it for some reason (which breaks cmake [1]). Dunno if that is a local thing here or if it is in svn.

        Martin

        [1]
        CMake Error: Error in cmake code at
        bmesh/blender/source/creator/CMakeLists.txt:110:
        Parse error. Expected a command name, got unquoted argument with text “<<<<<<<".
        — Configuring incomplete, errors occurred!

      • 22. joeedh  |  November 7, 2009 at 12:28 pm

        I have got to get cmake fixed. Anyway, easiest way to figure out compile problems is to show up on irc://irc.freenode.net/blendercoders and ask for help.

      • 23. oruchuban  |  November 6, 2009 at 9:31 pm

        Try this build http://www.graphicall.org/builds/builds/showbuild.php?action=show&id=1114 – works for me as a charm.

  • 24. Weelian  |  November 6, 2009 at 11:38 am

    Excellent! I’m going to try it out now!

    Reply
  • 25. Volant  |  November 6, 2009 at 1:28 pm

    The snap tool (Shift + S) is crashing almost every time for me. I use Cursor -> Selection a lot, so it’s annoying. 😦

    Reply
    • 26. joeedh  |  November 6, 2009 at 1:34 pm

      ok I’ll take a look

      Reply
      • 27. Volant  |  November 6, 2009 at 1:38 pm

        Thank you. I appreciate the fast reply!

  • 28. Martijn  |  November 6, 2009 at 3:48 pm

    1) Create a cube
    2) Select any edge ring
    3) CTRL-R Roll mouse wheel to make multiple cuts

    Bug: The purple preview lines are not perpendicular, instead they all cross each other.

    When I click to commit, however, the cuts ARE perpendicular and exactly the way as one would expect.

    Reply
    • 29. AkoZ  |  November 9, 2009 at 4:34 pm

      confirmed , just as this … just have a look on edge dividing to be in the right cut.

      Reply
  • 30. The Oblivious Prattler  |  November 7, 2009 at 4:25 am

    I noticed something odd, but it may not be a bug. If you subdivide a face (FC1) that’s adjacent to a face that has previously been subdivided (FC2), then the subdivision occurs by placing vertices in the centers of the shared and adjacent edges. There are also no edge marks showing that FC1 has been subdivided in any way. Does that make sense? Probably not. =(

    Reply
  • 31. Pottsdie  |  February 22, 2012 at 10:33 pm

    Link to the bug tracker is dead.

    Reply
  • 32. mattmerk  |  March 12, 2012 at 1:08 am

    Would love a Mac build to test. I have begun using Blender for VFX on a few TV shows. Modeling is a real weak point in the currently available Blender and I’ve really been looking forward to bmesh.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Trackback this post  |  Subscribe to the comments via RSS Feed


Categories

Feeds


%d bloggers like this: